1902] METEOROLOGICAL SCREEN 203 



to heart. I do not mean to imply that our education was 

 complete— as a matter of fact, we never ceased to learn new 

 tips or to adopt new devices, and the general sledging work 

 of the second summer was vastly superior to that of the first 

 — but it was the crushing ineffectiveness of our early efforts 

 which taught us the first great lesson. 



The daylight hours were now getting rapidly shorter, and 

 we knew that before the end of the month we should lose the 

 sun. We were left with little time to complete all our outside 

 arrangements, which had been necessarily delayed until the 

 formation of the ice-sheet ; although we felt anything but 

 certain that the ice had come to stay, the losses which its 

 break-up would entail must now be risked. 



One of our first cares was to get up the meteorological 

 screen ; this erection, made under the superintendence of our 

 meteorologist, Mr. Royds, consisted of a framework supported 

 by four stout poles; special louvred box-screens were placed 

 high on this, and inside them were fitted the various ther- 

 mometric and hygrometric instruments, whilst the corner poles 

 were utilised for anemometers and wind vane. The whole of 

 this somewhat elaborate erection was placed about loo yards 

 astern of the ship, and consequently in a direction which 

 would be to windward of her with the prevalent south-easterly 

 winds. At first the actual screens were some eight feet above 

 the surface of the ice, but we soon found that our small bay 

 was a focus for driving snow, and after each storm the surface 

 was raised a foot or more and the comparative height of the 

 screens proportionately reduced ; once the whole structure had 

 to be dug up and moved for this reason, but this could not 

 be repeated often, and the net result was that the screens were 

 reduced to an average height of five or six feet above the 

 surface. We found it was quite time that these screens were 

 placed, as we were getting very inaccurate temperature readings 

 on board ; for instance, I find a note of one comparison made 

 about this time : ' The thermometer on the gunwale shows 

 — 20°, that in the screen on board 4° higher, and that in the 

 screen on the ice some 5° or 6^ lower.' 



